Text (LLM), image, video, and multimodal—alongside the combined Model board.
Model leaderboard (all)
Rendimiento entre tareas (multimodal/visión/lenguaje; datos de ejemplo).
Cross-task overview; multimodal, vision, and language may split into additional columns or child boards once eval JSON is wired.
Public ranking policy: rows are sorted by composite score (desc). Composite score is a weighted sum of normalized sub-metrics; ties are broken by higher recent activity.
| Puesto | Modelo | Proveedor / equipo | Tipo | Puntuación | Notas |
|---|---|---|---|---|---|
| 1 | Demo-Vision-Pro | Demo Lab | Multimodal | 94.2 | Equilibrio imagen + texto |
| 2 | NorthStar-MM | North AI | Multimodal | 92.8 | Fuerte en contexto largo |
| 3 | Aurora-VL-7B | Aurora | Visión-lenguaje | 91.5 | Adecuado para edge |
| 4 | Helix-3 | Helix Research | General | 90.1 | Llamadas a herramientas estables |
| 5 | Kite-Small | Kite | Lenguaje | 88.6 | Buena relación calidad/precio |
| 6 | Lattice-R1 | Lattice | Razonamiento | 87.9 | Subpuntuaciones fuertes en mates/código |
| 7 | Pulse-Audio-2 | Pulse | Multimodal de voz | 86.4 | ASR/TTS combinados |
| 8 | Quark-Mini | Quark Systems | Lenguaje | 85.2 | Baja latencia |